Leishmania donovani-infected splenic macrophages and P388D1 (P388D1(I)) failed to activate T cells in response to low dose of exogenous peptide. The membrane fluidity of P388D1(I) was greater than that of the normal counterpart P388D1(N), but could be reduced either by exposing the cell below phase transition point or by loading cholesterol into membrane (L-P388D1(I)), and this was associated with enhanced Ag-presenting ability of P388D1(I). Presentation of endogenous leishmanial Ag, kinetoplastid membrane protein-11, was also defective, but could be corrected by loading cholesterol into membrane. Because membrane rafts are important for Ag presentation at a low peptide dose, raft architecture of P388D1(I) was studied using raft (CD48 and cholera toxin-B) and non-raft (CD71) markers in terms of their colocalization with I-Ad. Binding of anti-CD48 mAb and cholera toxin B subunit decreased significantly in P388D1(I), and consequently, colocalization with I-Ad was not seen, but this could be restored in L-P388D1(I). Conversely, colocalization between I-Ad and CD71 remained unaffected regardless of the presence or the absence of intracellular parasites. P388D1(N) and L-P388D1(I), but not P388D1(I), formed peptide-dependent synapse with T cells quite efficiently and this was found to be corroborated with both intracellular Ca2+ mobilization in T cells and IL-2 production. This indicated that intracellular parasites disrupt the membrane rafts, possibly by increasing the membrane fluidity, which could be corrected by making the membrane rigid. This may be a strategy that intracellular L. donovani adopts to evade host immune system.
Leishmaniasis is a vector-borne parasitic disease with a broad range of clinical manifestation from local cutaneous lesion to life-threatening visceral disease, mainly caused by differences among Leishmania species and the immunological status of the mammalian hosts (1). The parasite exists in two different stages; the flagellated promastigotes are transmitted with the bite of the sandfly to the mammalian hosts, where they transform into the amastigote stage. Although macrophage (Mφ)4 are generally considered to be the main or only host cells allowing Leishmania to survive and multiply (2), other cells can harbor intact viable parasites at least transiently. For example, Leishmania major parasites have been observed in polymorphonuclear leukocytes, fibroblast, and dendritic cells (DC) at different stages of infection (3). The ability of Langerhans cells and DCs to take up Leishmania is still a matter of debate (4, 5). Murine fetal skin-derived DCs are preferentially infected by L. major amastigotes compared with metacyclic promastigotes (5). Recently, studies show that human blood monocyte-derived DCs could be infected with Leishmania donovani (LD) (6), and maturation of mouse bone marrow-derived DCs is inhibited by Leishmania amazonensis (7). The present study is concerned with the Ag-presenting ability of Mφ as a general mechanism to understand how intracellular Leishmania parasites modulate the surface phenomena of APCs, which, in turn, may affect Ag presentation.
The disease visceral leishmaniasis or kala-azar, caused by the protozoan parasite LD, is characterized by defective cell-mediated immunity (8, 9, 10, 11, 12, 13), the cause of which is still unknown. Leishmania-infected Mφ are unable to present even processing-independent peptide sequences to T cell hybridoma, and this inability is not due to defective MHC expression (14). Recent studies show that Leishmania-infected Mφ efficiently stimulate Ag-independent and Ag-dependent TCR-mediated capping of F-actin in T cells (15). Infected Mφ are less efficient at promoting the sustained T cell signaling events, but this is not due to altered levels of surface-expressed peptide-MHC complexes (15).
Physiological stimulation of T cells by antigenic peptides presented in the context of MHC molecules requires a stable contact between T cells and APCs to enable the generation of efficient activation signals (16, 17). The area of contact between the two cells, the immunological synapse, has recently become a fruitful focus of investigation (18, 19). Recent work on the membrane features that affect presentation of Ags by APCs to T cells indicates that lipid rafts in the outer leaflet of the membrane lipid bilayer induce clustering of the peptide-MHC complex on the APC surface. This leads to an effective increase in the surface density of the MHC-Ag and up-regulation of the T cell response (20, 21, 22). Lipid rafts are cholesterol- or sphingolipid-rich microdomains with high melting temperatures caused by ordered packing of molecules (23). The proposed functions of lipid rafts include cholesterol transport (24, 25), endocytosis (26), signal transduction (27, 28, 29), potocytosis (30), and pathogen portals (31, 32, 33, 34). The concept of lipid rafts as pathogen portals has gained considerable interest in recent years. When Plasmodium invades RBC, host cell plasma membrane lipids are incorporated into the parasite-containing vacuole, whereas transmembrane proteins are excluded (35). Similarly, in Toxoplasma gondii infection, GPI-anchored surface proteins are incorporated into parasitophorous vacuole, whereas host cell transmembrane proteins are excluded from it (36). Cholesterol appears to play a key role in organizing the lipid domain, which is required for the entry of mycobacteria (37) and Leishmania (38) into the host cells. Disruption of lipid rafts by sequestration of cholesterol after treatment of APCs with methyl β-cyclodextrin impaired the presentation of peptide-MHC complexes (22). The effect could be reversed at high Ag concentrations, indicating that lipid rafts indeed play an important role in enhancing the T cell response at low ligand concentrations (39). Interestingly, changes in the membrane fluidity of cells occur under a variety of pathological conditions (40, 41).
Such changes in fluidity are reflected in reduced T cell-mediated cytotoxicity (42, 43, 44) as well as NK cell-mediated target cell lysis (45). Previously, we have demonstrated that a subset of splenic Mφ of Leishmania-infected BALB/c mice display an increase in membrane fluidity coupled with defective Ag-presenting function; this inability to present Ag was not due to the lack of peptide-MHC complex formation on the surface of Mφ (11). The present work demonstrates for the first time that during their intracellular life cycle, Leishmania parasites disrupt the membrane raft of Mφ by increasing membrane fluidity. This increase in fluidity could be corrected either by exposing the parasitized Mφ to lower than phase transition temperature or by liposomal delivery of cholesterol, leading to correction in Ag presentation. Thus, Leishmania may evade the host immune response by exploiting rafts during their intracellular lifecycle.
Materials and Methods
Abs and other reagents
FITC-conjugated anti-mouse I-Ad, PE-conjugated mAb to mouse CD71 and CD48, biotin-conjugated anti I-Ad, and streptavidin-PE were purchased from BD Pharmingen. FBS, the peptide LEDARRLAIYEKK (N-terminal 12–26 aa residues of λ repressor protein, also defined as λR12–26), and RPMI 1640 medium were obtained from Invitrogen Life Technologies. Penicillin-streptomycin, sodium bicarbonate, HEPES, 2-ME, phosphatidylethanolamine, 1,6-diphenyl-1,3,5-hexatriene (DPH), fura 2-AM, paraformaldehyde, and FITC-conjugated cholera toxin B subunit (CTX-B-FITC) were purchased from Sigma-Aldrich. Cholesterol was obtained from Merck. [3H]TdR (sp. act., 6.7 Ci/mmol) was purchased from New England Nuclear. Na251CrO4 was obtained from Bhaba Atomic Research Center (sp. act., 65.1Ci/g). Hen egg lysozyme (HEL) was obtained from Roche. Ab C4H3 specific for IAk-HEL46–61 (46) was a gift from Dr. R. N. Germain (National Institutes of Health, Bethesda, MD). Ab C4H3 was conjugated to FITC as previously described (47).
Murine Mφ-like tumor cell P388D1 (48, 49) was used in in vitro experiments. T cell hybridoma 9H3.5 (50) of BALB/c origin, specific for λR12–26, was a gift from Prof. M. L. Gefter (Massachusetts Institute of Technology, Cambridge, MA). The IL-2-dependent cell line HT-2 was obtained from American Type Culture Collection. All cells were maintained in RPMI 1640 medium supplemented with 10% FCS and 2-ME (5 × 10−5 M) at 37°C with 5% CO2 in a humidified atmosphere.
Parasite maintenance and preparation of soluble leishmanial Ag (SLA)
LD strain AG83 (MHOM/IN/1983/AG83), originally obtained from an Indian kala-azar patient (51), was maintained in Golden hamsters as described previously (52). Promastigotes obtained after transforming amastigotes from the spleen of infected animals were maintained in culture in medium 199 (Invitrogen Life Technologies) supplemented with 10% FCS at 22°C. The culture was replenished with fresh medium every 72 h. SLA was prepared as described previously (51).
Mice, infection, and preparation of splenic and peritoneal Mφ
BALB/c mice were obtained from the animal facility of the institute. CBA/J mice were a gift from Dr. S. K. Basu (National Institute of Immunology, New Delhi, India). Mice were housed under conventional conditions, with food and water ad libitum. Animals were used for experimental purposes with prior approval of the institutional animal ethics committee. Spleens of BALB/c mice were isolated and macerated between the frosted ends of a pair of glass slides in PBS (pH 7.2). The cell suspension was then layered over Ficoll-Hypaque, and density gradient centrifugation was performed at 1600 rpm for 20 min. The cells in the interface were collected and washed twice in PBS. The splenic Mφ were purified as previously described (12). Similarly, peritoneal Mφ of CBA/J mice were isolated as previously described (53). For infection, 6- to 8-wk-old BALB/c mice were inoculated with 107 early passage promastigotes via the intracardiac route (51).
Splenic Mφ and P388D1 cell lines were used for in vitro infection studies. Cells (105) were allowed to adhere on glass coverslips for 3 h at 37°C in presence of a 5% CO2 atmosphere, after which the nonadherent cells were removed by gentle washing with serum-free medium. The adherent Mφ, after overnight incubation in complete medium, were challenged with LD promastigotes at an Mφ to parasite ratio of 1:10 and incubated further for 6 h at 37°C. Excess parasites were then washed off with serum-free medium. The Mφ were then incubated further for 6, 12, 24, 36, 48, 60, and 72 h to determine their ability to support intracellular parasite replication (54). The splenic Mφ and P388D1, infected or not, were used as APCs to drive T cell hybridomas in the presence or the absence of λR12–26 (55).
Ag presentation and peptide pulsing
The APCs were incubated for 24 h with specific peptide and T cell hybridoma in complete RPMI 1640 medium in a 37°C incubator. The culture supernatants were analyzed for the presence of IL-2 by growing an IL-2-dependent cell line, HT-2, in the supernatants. HT-2 (104/well) was incubated with a 50% concentration of culture supernatant for 24 h. The cells were then pulsed with 1 μCi of [3H]thymidine for the last 18 h (55). The incorporation of radioactive thymidine was assessed by a scintillation counter (Packard). Uninfected P388D1 (P388D1(N)) and infected P388D1 (P388D1(I)) cells (106/ml) were pulsed with 2 μM λR12–26 peptide for 1 h at 37°C and extensively washed with serum-free RPMI 1640 medium. They were then resuspended in complete RPMI 1640 medium to a cell density of 2 × 105/ml. These peptide-pulsed P388D1(N) and P388D1(I) cells were used as APCs to drive peptide-specific T cell hybridomas. Normal Mφ were pulsed with HEL (1 mg/ml) for 4 h, then washed and kept for another 24 h. Finally, cells were stained with FITC-labeled-C4H3, and the extent of peptide-MHC complex present on the cell surface was determined by FACS analysis (15).
Measurement of fluorescence anisotropy (FA)
The membrane fluorescence and lipid fluidity of cells were measured following the method described by Shinitzky and Inbar (56). Briefly, the fluorescent probe DPH was dissolved in tetrahydrofuran at 2 mM concentration. To 10 ml of rapidly stirring PBS (pH 7.2), 2 mM DPH solution was added. For labeling, 106cells were mixed with an equal volume of DPH in PBS (Cf 1 μM) and incubated for 2 h at 37°C. Thereafter the cells were washed thrice and resuspended in PBS. The DPH probe bound to the membrane of the cell was excited at 365 nm and the intensity of emission was recorded at 430 nm in a spectrofluorometer. The FA value was calculated using the equation: FA = [(I∥ − I⊥)/(I∥ + 2I⊥)], where I∥ and I⊥ are the fluorescent intensities oriented, respectively, parallel and perpendicular to the direction of polarization of the excited light (57).
Phase transition in relation to Ag presentation
In one set, P388D1(N) and P388D1(I) cells were pulsed with 2 μM peptide for 1 h, then washed and resuspended in complete RPMI 1640 at a final cell density of 106 cells/ml. The cells were incubated at 10, 15, 20, 25, 30, or 37°C for 15 min and then fixed in 1% paraformaldehyde in PBS. They were seeded in a 96-well microtiter plate (Falcon) at a concentration of 2 × 104 cells/well, incubated with 9H3.5 T cell hybridoma (105 cells/well) for 24 h at 37°C, and the IL-2 produced was measured by a previously described method (55). P388D1(I) cells were pulsed with the peptide at 37°C, then exposed to 15°C for 15 min and fixed with paraformaldehyde (prepulsed). In another situation, P388D1(I) cells were exposed to 15°C for 15 min, then fixed with paraformaldehyde and pulsed with peptide (postpulsed).
Liposome preparation and delivery of cholesterol
Liposomes were prepared with cholesterol and phosphatidylethanolamine at a molar ratio of 1.5:1 as previously described (58, 59). Briefly, a thin dry film of lipids (5.8 mg cholesterol and 8.0 mg phosphatidylethanolamine) was dispersed in 1 ml of RPMI 1640 and sonicated at 4°C three times, 1 min each, at maximum output. To alter the fluidity of cells, 105 intact cells were incubated with liposomes for 12 h at 37°C. The cells with altered fluidity were then washed three times in serum-free RPMI 1640 medium and finally resuspended in 10% FCS containing RPMI 1640. The FA value of the liposome-treated cells was calculated as previously described (57).
Analysis of peptide-MHC complex
Peritoneal Mφ of CBA/J mice were infected with LD for 6 h, washed, and kept overnight. In another set, infected Mφ after 6 h of infection were treated with cholesterol-rich liposome for overnight. Normal, infected, and cholesterol-loaded infected Mφ were pulsed with HEL (1 mg/ml) for 4 h, washed, kept overnight, and stained with C4H3-FITC conjugate for 30 min. In another set, peptide-pulsed infected splenic Mφ (I-Mφ) were exposed to 15°C for 15 min (prepulse), stained with C4H3-FITC at 15°C for 30 min, and fixed with paraformaldehyde in a chilled condition. The extent of binding of C4H3-FITC was analyzed by flow cytometry.
Confocal microscopy and analysis of conjugate formation
P388D1(N), P388D1(I), and cholesterol-loaded P388D1(I) (L-P388D1(I)) cells were harvested, washed, and resuspended in cold wash buffer (PBS/0.1% NaN3/1% FBS), centrifuged at 350 × g for 5 min, and finally resuspended in 50 μl of wash buffer. Cells were stained with fluorochrome-conjugated Ab according to the manufacturer’s protocol. Briefly, FITC-conjugated anti I-Ad was diluted to a predetermined optimal concentration in 50 μl of FACS buffer. Cells (106) in 50 μl of wash buffer were added to diluted conjugate, mixed thoroughly by tapping, and incubated at 4°C in the dark. The reaction was stopped by adding 200 μl of FACS buffer and washed three times. For staining with second Ab, PE-conjugated anti-CD48/CD71 was added as described above. In another set, the above cell types (P388D1(N), P388D1(I), and l-P388D1(I)) were stained with biotinylated anti I-Ad and streptavidin-PE, then costained with FITC-CTX-B following the protocol described above. For studying the conjugate formation, APCs (P388D1(N), P388D1(I), and L-P388D1(I)), either unpulsed or pulsed with 2 μM λR12–26 for 1 h at 37°C, were labeled with biotinylated anti I-Ad and then with streptavidin-PE (19). The T cell hybridomas were stained with CTX-B-FITC. APC and T cells, mixed at a ratio of 1:5, were resuspended in FACS buffer and allowed to form conjugate for 30 min. The cells were then fixed with 1% paraformaldehyde, mounted with 90% glycerol on a glass slide, and observed under a laser scanning microscope (LSM 510; Zeiss). Cells found in joint couplets under phase contrast and exhibiting a zone of colocalization under confocal were considered to have formed a synapse. The number of such synapse-forming couplets per 100 APCs is presented as percentage of synapse formation.
Anti-kinetoplasid membrane protein-11 (anti-KMP-11) CTL response
KMP-11-expressing mammalian expression vector construct (pCMV-LIC KMP-11) used for immunization purpose was generated as previously described (60). Nonadherent splenocytes from pCMV-LIC KMP-11 DNA-vaccinated BALB/c mice (3 wk after immunization with two booster doses of 100 μg of plasmid, 7 days apart i.m.) stimulated with SLA for 4 wk were used as effectors. The targets were the splenic Mφ (106) of normal and 2-month infected BALB/c mice either treated or not with liposome. These were labeled with 51Cr (100 μCi) for 1 h at 37°C in 5% CO2 incubator and washed several times until no gamma irradiation was detected in the supernatant. The effectors and targets were mixed in round-bottomed, 96-well plates (200 μl) at various (12:1, 25:1, and 50:1) ratios. After 4-h incubation, 100 μl of culture supernatant was collected and counted in triplicate in a liquid scintillation counter (Tri-Carb 2100TR; Packard). Specific lysis was calculated according to the formula: % specific lysis = (sample − spontaneous release)/(maximum release − spontaneous release) × 100 (60).
Intracellular calcium mobilization in T cells
Intracellular Ca2+ mobilization in T cell hybridoma was monitored in response to peptide-pulsed and unpulsed APCs. Briefly, T cell hybridomas (106/ml) were taken in RPMI 1640 (without FCS), washed once, and resuspended in RPMI 1640 containing 6 μM fura 2-AM for 60 min at 37°C in the dark with gentle shaking. The cells were washed with HBSS and resuspended in HBSS containing 0.5 μM EGTA at a cell number of 106/ml. The cell suspension (1 ml) was placed in a continuously stirred cuvette at room temperature in a fluorometer (Hitachi U440). Fluorescence was monitored in real time at λex1 = 340 nm, λex2 = 380 nm, and λem = 510 nm at a bandwidth of 10 nm, and the data were presented as the relative ratio of fluorescence excited at 340 and 380 nm. After ∼1 min of scanning, the peptide-pulsed or unpulsed APC (P388D1(N), P388D1(I), and L-P388D1(I)) were added to the cuvette, and the ratio of fluorescence was monitored (61).
Statistical variation and presentation
Each experiment was performed three to five times, and representative data from one set of these experiments are presented; the interassay variation was within 10%. Results are expressed as the mean ± SD of the individual set of experiments.
Infection of splenic Mφ and P388D1 cells with LD
Splenic Mφ were infected with LD, and intracellular parasites were enumerated after 6, 12, 24, 36, 48, 60, and 72 h of infection. After 24 h of infection, 80% of splenic Mφ were found to be infected harboring ∼4 parasites per Mφ (Fig 1). The maximum number of I-Mφ was observed at 48 h of infection, and at this point the number of parasites per Mφ was six (Fig. 1). P388D1, a mouse Mφ-like cell, supports Leishmania replication and has been in use over the years as a model host instead of primary Mφ culture (49). Enumeration of the number of internalized parasites in P388D1(I) cells after 6, 12, 24, 36, 48, 60, and 72 h of infection showed that the intracellular parasite count increased up to 48 h. At 24 h, >95% of the cells were infected. The percentage increased slightly at 48 h and then reached a plateau (Fig. 1). At 24 and 48 h, the average numbers of intracellular parasites were 10–12 and 12–14, respectively. In the splenic Mφ, maximum parasite number was attained at 48 h; in the case of P388D1(I) cells, near-maximal internalization of parasites was attained at 24 h. After this, studies were conducted with 48-h infected splenic Mφ and 24-h infected P388D1 cells as APC.
Inability of LD-infected Mφ and P388D1 to present peptide Ag at a low peptide dose
The Ag-presenting ability of splenic Mφ infected with LD was investigated with increasing concentrations of λR12–26. Because our study is concerned with cell surface phenomena associated with Ag presentation, we have used the processing-independent peptide sequence λR12–26 to drive T cells (14). I-Mφ failed to activate 9H3.5 at a low peptide concentration, unlike normal splenic Mφ. This inability could largely be overcome by loading the APC with a high dose of the peptide (10 μM λR12–26; Fig. 2). Identical observations were made using P388D1(N) and P388D1(I) as APCs (Fig. 3). In some of the subsequent experiments, we pulsed P388D1(N) and P388D1(I) cells with λR12–26 peptide. Ag presentation assays using these peptide-pulsed APCs showed results essentially similar to those of assays in which peptide was kept throughout in the culture medium (Fig. 3, inset).
Membrane fluidity under parasitized condition
Leishmania-infected Mφ are reported to be unable to present processing-independent peptide to T cell hybridoma (14), and changes in membrane fluidity in effector cells influence T cell-mediated cytotoxicity (43). Because LD-infected splenic Mφ and P388D1 showed identical patterns of response in terms of Ag presentation, the rest of the studies were conducted with P388D1 cells. Therefore, we studied the membrane fluidity of P388D1(N) and P388D1(I) cells in terms of FA using DPH as probe. The observed decrease in FA (0.35 vs 0.23) due to infection at 24 h (Fig. 4) indicated an increase in membrane fluidity.
Influence of membrane fluidity on Ag presentation
If an increase in membrane fluidity in infected cells was indeed responsible for defective Ag presentation, the efficiency of Ag presentation by normal APCs would be expected to undergo a sharp change within a narrow temperature range characteristic of the phase transition of the membrane lipid bilayer (23). P388D1(N) cells were therefore pulsed with 2 μM peptide, and aliquots of cells were exposed to various temperatures, i.e., 10, 15, 20, 25, 30, and 37°C for 15 min. These cells were then fixed with 1% paraformaldehyde and used as APCs. Corresponding FA values of fixed APCs were calculated using DPH as a probe. There was a concomitant increase in the FA when cells were exposed to a lower temperature, indicating a decrease in membrane fluidity (Fig. 5). Interestingly, Ag presentation was enhanced ∼2-fold as the incubation temperature was decreased from 20 to 15°C (Fig. 5). The results presented above suggest that there is a significant enhancement in the potency of APC function when cells are chilled. The potency difference may result from better APC function upon chilling, and this should be reflected in the APC dose-response graph with a fixed number of T cells. The maximal T cell activation was observed with 2 × 105 peptide-pulsed P388D1(N) exposed to 37°C. Interestingly, to obtain an identical response, only 2 × 103 of 15°C-exposed, peptide-pulsed P388D1(N) were required (Fig. 5, inset). When the results were expressed as the fold enhancement in Ag-presenting ability between the cells exposed to 15 vs 37°C for a given cell number, this ratio was found to vary between 2.05 and 2.8. The transition temperature between the two states was between 20 and 15°C (Fig. 5). This indicated a clear inverse relationship between membrane fluidity and Ag presentation. Henceforth, the cells exposed to a below phase transition temperature were defined as chilled cells.
To study the membrane turnover, if any, that can contribute to defective APC function during infection, P388D1(I) cells were pulsed with peptides either before (prepulse) or after (postpulse) fixing with paraformaldehyde. As usual, P388D1(I) failed to induce T cell activation. Interestingly, prepulsing and chilling P388D1(I) caused a significant enhancement of Ag presentation (Fig. 6). An essentially identical response was obtained when postpulsed and chilled P388D1(I) were used (Fig. 6). When the results were expressed as the fold enhancement in Ag-presenting ability between the cells exposed to 15 or 37°C for a given cell number, this ratio varied between 6.3 and 10 in the case of P388D1(I). Because a decrease in FA due to infection (Fig. 4) is essentially equal to an increase in FA upon chilling (Fig. 5), it was relevant to know whether chilled infected APC would behave as normal APC at a low peptide dose to drive T cells. In this event, it was observed that at a low peptide dose, chilled P388D1(I) behaved like P388D1(N). P388D1(I) that were not chilled failed to stimulate T cells, as expected (Fig. 7).
There are a number of reports stating that fluidity can also be decreased by incorporating cholesterol in the membrane (62, 63, 64, 65). Therefore, we were interested in making P388D1 membrane rigid by liposomal delivery of cholesterol. Because infection of the P388D1 cell line leads to an increase in membrane fluidity (Fig. 4), as seen in the case of cholesterol extraction in other systems (66, 67), we tried to incorporate cholesterol in the membrane of P388D1(I) by liposomal delivery to make the membrane rigid and to study the cell’s Ag-presenting function. Cholesterol was delivered through the liposome to infected cells. This resulted in an increase in the FA value of these cells, i.e., the fluidity decreased compared with cholesterol-undelivered cells, making it comparable to that of normal cells (Fig. 8). The L-P388D1(I) cells showed ∼3-fold enhancement of Ag-presenting ability (Fig. 8). Thus, a decrease in the fluidity of P388D1(I) cells by loading cholesterol almost corrected the defective Ag-presenting ability.
Quantification of cell surface peptide-MHC complex
It has been reported that surface MHC class II molecules are comparable in normal as well as parasitized Mφ (15). Because Abs that can detect the λ-repressor-peptide I-Ad complex are not available, we studied the expression of the I-Ak-HEL46–61 complex in peritoneal Mφ of CBA/J mice using the specific mAb, C4H3 (15, 46). Normal and infected Mφ were then pulsed with HEL and fixed with paraformaldehyde, and the peptide-MHC complex was quantitated with C4H3-FITC. LD infection did not inhibit cell surface expression of the peptide-MHC complex. Similarly, there was not much difference in the binding of C4H3-FITC compared with HEL-pulsed normal Mφ. HEL-pulsed infected Mφ exposed to 15°C or treated with cholesterol-rich liposome showed essentially similar binding with C4H3-FITC as HEL-pulsed normal Mφ. This indicated that an essentially identical level of peptide-MHC complex was present on the Mφ surface (Fig. 9).
Presentation of leishmanial Ag
We also studied the ability of a cell surface-associated parasite Ag of intracellular origin taken from 2-mo-infected mice as a target of antileishmanial T cells. Because KMP-11 is expressed on the surface of the splenic Mφ of infected mice (data not shown), anti-KMP-11 T cells were generated by priming mice with the KMP-11 DNA construct, and a CTL assay was performed using splenic Mφ from 2-mo-infected mice as targets. There was no lysis when Mφ from normal animal were used as targets (60). Using splenic Mφ from infected mice at an E:T cell ratio of 25:1, there was ∼16.5% lysis. In contrast, when the splenic Mφ of infected mice were treated with liposomal delivery of cholesterol and then used as a target, there was 36.2% lysis at the above E:T cell ratio. This indicated that Ags derived from intracellular parasites are not presented well due to an increase in the fluidity of parasitized Mφ (Fig. 10).
Leishmania infection disrupts lipid rafts, which reappear after cholesterol delivery to infected cells
Because liposomal delivery of cholesterol to P388D1(I) cells decreased membrane fluidity and enhanced Ag-presenting ability, and because cholesterol is known to partition between raft and nonraft phases (68, 69), we decided to investigate the raft architecture under the parasitized condition. For this study, CD48/CTX-B and CD71 were used as raft (70, 71) and non-raft (72) markers, respectively. We presumed that disruption of raft under parasitized condition would prevent colocalization between CD48/CTX-B and I-Ad, whereas colocalization between CD71 and I-Ad would remain unaffected. We studied the colocalization between CD48/CTX-B and I-Ad and also between CD71 and I-Ad. From the confocal images, it was apparent that there was no down-regulation of I-Ad expression in the above conditions (Fig. 11,A). Staining of the three sets of cells (P388D1(N), P388D1(I), and l-P388D1(I)) with anti-CD48 showed that P388D1(N) and L-P388D1(I), but not P388D1(I), stained well (Fig. 11,A). Colocalization studies between CD48 and I-Ad showed that this occurred only with P388D1(N) and L-P388D1(I) cells, not with P388D1(I) cells (Fig. 11 A).
Identical studies were performed in I-Ad and CD71 to show that I-Ad could colocalize with nonraft markers. Colocalization between the above markers was observed in all three conditions (Fig. 11,A). To confirm that the reduced expression of CD48 is not a unique case, we decided to use another raft marker as a probe. The ubiquitously expressed glycosphingolipid ganglioside, GM1, which is highly expressed in lipid rafts and CTX-B, binds selectively with the external components of GM1 (71). Therefore, we used CTX-B-FITC as a probe to assess the status of rafts in the above conditions. Under parasitized conditions, there was a significant reduction of CTX-B-FITC binding, which could be restored upon cholesterol loading. In this study we also observed colocalization between I-Ad and CTX-B in the case of L-P388D1(I) cells (Fig. 11 B). The above observations support the idea that Leishmania infection disrupts raft formation, which can be restored upon liposomal delivery of cholesterol.
Formation of APC-T cell conjugates and intracellular Ca2+ mobilization
The ability of peptide-unpulsed and -pulsed P388D1(N), P388D1(I), and L-P388D1(I) cells to form synapses with the peptide-specific T cell hybrid was studied. Optimal synapse formation was observed at ∼30 min of incubation of APC-T cells. Peptide pulsing led to efficient conjugate formation between 9H3.5 and P388D1(N)/L-P388D1(I), but not with P388D1(I), and the extent of conjugate formation was 75, 77, and 15%, respectively. Thus, synapse formation is severely compromised with P388D1(I), and this can be corrected by loading cholesterol. In the absence of the peptide, a basal level of conjugate formation was observed (∼2%). The picture was somewhat similar when intracellular Ca2+ mobilization in the T cell was studied in response to peptide-pulsed and -unpulsed APCs. Intracellular Ca2+ mobilization was observed when T cells were stimulated with peptide-pulsed P388D1(N) and L-P388D1(I). Interestingly, only a slight mobilization occurred when stimulation was performed with P388D1(I) (Fig. 12). There was no mobilization when T cells were stimulated with unpulsed APCs.
Our study clearly showed that I-Mφ and P388D1(I) decreased the magnitude of T cell activation at a low dose of peptide, but both responses were essentially identical with that of their normal counterpart at a high peptide dose. Our study was concerned with cell surface phenomena of Ag presentation; therefore, we used a processing-independent peptide sequence (λR12–26) to activate T cells (14). Because parasitized Mφ are capable of activating T cells only at a high peptide dose, it is possible that during the active stage of the disease, when there is a high level of circulating parasite Ag, there may not be any defect in Ag presentation. Identification of free leishmanial Ag(s) in the circulation was not possible on many occasions even in confirmed kala-azar patients (73). Similarly, detection of free uncomplexed Ag in the circulation of cotton rats infected with LD has not been successful (74).
In many systems the presence of specific Ab impedes the detection of free Ag in serum due to the formation of immune complexes (74). Previously we demonstrated that a subset of splenic Mφ of LD-infected BALB/c mice showed increased membrane fluidity and defective Ag-presenting ability (11). The question arises of whether the rigidity of the membrane lipid bilayer is a critical determinant of the Ag-presenting ability of Mφ in vitro. The increase in membrane fluidity in P388D1(I) was indicated by a significant decrease in FA. However, the inverse correlation between membrane fluidity and Ag presentation might be a coincidence without mechanistic significance. To explore this possibility, we examined the effect of a physical perturbation, such as temperature, on the Ag-presenting ability of P388D cells. Such an effect is likely to be more amenable to interpretation than that caused by infection. For this purpose, the normal and infected cells were exposed to a temperature of 37°C, the physiological temperature at which the bulk of the lipid bilayer is in a fluid, liquid crystalline state, allowing lateral mobility of membrane components, and to 15°C, a temperature at which the bilayer, especially the cholesterol-sphingolipid domain rich in saturated acyl chains, exists in an ordered, solid, gel-like state, causing severe restriction in lateral mobility (75). With P388D1(N), we observed a significant decrease in membrane fluidity at 15°C, as indicated by an increase in FA, coupled with a 2-fold increase in Ag-presenting ability. The effect of incubation at low temperature was even more marked with P388D1(I); the enhancement in Ag-presenting ability was 6.3- to 10-fold. The mechanism of enhanced Ag presentation by chilled APCs is not clear. There is a report that conjugate formation between T cells (influenza virus hemagglutinin317–329 peptide specific) and APC (A20 B cells) is marginally increased by exposing APCs to 5°C (76). Thus, a decrease in FA is essentially equal to its increase upon chilling, and the T cell responses of P388D1(N) and chilled P388D1(I) at a low peptide dose were also identical. This association lends credence to the idea that an increase in membrane fluidity may be one of the important underlying mechanisms of the defective T cell response in leishmaniasis.
The inability to present peptide was not due to membrane turnover, because an identical response was obtained if P388D1(I) was fixed either before pulsing with peptide (postpulsing) or after (prepulsing). To determine whether a peptide-MHC complex is indeed present on the cell surface, we analyzed this complex. Because there is no specific Ab available to quantitate the I-Ad-λR12–26 complex, we used C4H3, an Ab specific for I-Ak-HEL46–61 (46). Bindings of C4H3 to HEL-pulsed I-Mφ and chilled I-Mφ were comparable to its binding to normal Mφ. This observation reinforced the above finding that a stable peptide-MHC complex was formed when Mφ were infected. A similar observation has been reported by another group (15). Because loading of cholesterol is more physiological to modulate the fluidity of biological membrane (68) and steady state FA in model membrane is a sensitive measure of cholesterol dependent ordering (77), we studied the effect of cholesterol loading on Ag presentation. The lipid order detected by DPH-PC anisotropy in plasma membrane vesicle is highly sensitive to the amount of cholesterol, further strengthening its utility as a quantitative indicator of lipid raft content within fluid membranes (77). We therefore made use of liposomal delivery of cholesterol to make the membrane rigid. There is a report that increase in cholesterol in the extracellular milieu may augment Ag presentation by modulating the expression of HLA-D region products on APCs (69). Loading of cholesterol in P388D1(I) indeed decreased the membrane fluidity with concomitant increase in Ag presenting ability.
We then asked whether the results observed with exogenously added Ag were true with endogenously derived parasite Ag of infected Mφ and studied the presentation of KMP-11 to anti-KMP-11 CTLs (60). KMP-11, a surface membrane protein expressed in both amastigote and promastigote forms of the parasite, is emerging as a potential vaccine candidate (60). Results from our study clearly demonstrated that splenic Mφ from infected BALB/c mice were specifically recognized and lysed by anti-KMP-11 T cells only when they were treated with cholesterol in vitro. This indicated that despite the presence of KMP-11 in infected Mφ, as studied by immunofluorescence using anti-KMP-11 Ab (data not shown), Ag presentation was rendered defective by altered membrane fluidity in pathological conditions, which can be reversed by cholesterol loading. It may be recalled in this connection that class I MHC molecules become raft associated under the influence of some pathogens (32). Because loading of cholesterol in the infected Mφ enhances Ag presentation, it appears that infection leads to an increase in membrane fluidity, and this can be corrected by making the membrane rigid. The disruption of these microdomains or rafts by methyl-β-cyclodextrin treatment, which specifically extracts cholesterol, at a limiting concentration of Ag dramatically inhibits Ag presentation, which can be overcome by loading the APCs with exceptionally high doses of Ag (22). Thus, raft concentrates MHC class II molecules into microdomains, which allows efficient Ag presentation at low ligand densities (22).
The inability of P388D1(I) to present Ag at low ligand densities may be due to disruption of rafts brought about by intracellular LD parasites. Because liposomal delivery of cholesterol resulted in a decrease in membrane fluidity and enhancement of the Ag-presenting ability of P388D1(I) cells, we hypothesize that intracellular parasites lead to the disruption of rafts by an as yet unidentified mechanism that can be reversed upon cholesterol loading. Loading of cholesterol in infected Mφ showed similar binding of C4H3-FITC, which was comparable in infected Mφ, as in infected and chilled Mφ. This observation indicated that there was no overall change in the quantum of peptide-MHC complex. One of the explanations for the increased potency produced by either chilling or loading cholesterol is a change in the interaction between the peptide-MHC complex and TCR. Cholesterol is thought to serve as a spacer between the hydrocarbon chains of the sphingolipid and to function as a dynamic glue that maintains the raft assembly (68). In addition, cholesterol promotes the coexistence of solid, gel-like as well as fluid, liquid crystalline domains of the membrane and thereby prevents collapse of the membrane into either gel or fluid phases with a change in temperature (78). In other words, cholesterol moderates the effect of temperature shock. We may recall in this context that the lowering of temperature caused a significant increase in the Ag-presenting ability of infected cells compared with normal cells.
Thus, we studied the raft architecture in P388D1(N), P388D1(I), and L-P388D1(I). It is known that MHC class II molecules are both raft and nonraft associated (20). Using CD48 and CTX-B as raft markers (70, 71) and a transferrin receptor (CD71) as a nonraft marker (72), we studied the colocalization of I-Ad molecules. Curiously, both CD48-PE- and CTX-B-FITC stained P388D1(N) and L-P388D1(I) cells effectively, but not P388D1(I) cells. Interestingly, I-Ad molecules colocalized with CD71, but not with CD48/CTX-B, in P388D1(I), suggesting the disruption of rafts due to the presence of intracellular LD. The raft became evident again from the colocalization between I-Ad and CD48/CTX-B, when P388D1(I) cells were loaded with cholesterol. This suggests that the presence of intracellular parasites somehow disrupts the raft structure, which can be restored upon cholesterol loading. Naturally the question arises of which physical parameters of a biological membrane are influenced by cholesterol? Membrane cholesterol regulates membrane-embedded receptors in terms of their affinity state, the binding capacity and signal transduction (79). Most important, cholesterol may stabilize receptors in defined conformations related to their biological functions (79). Cellular cholesterol levels affect the shedding of IL-6R (80) and CD30 (81), which may be mediated by increased accessibility of the protease to the receptor cleavage site, and depletion of cholesterol from the plasma membrane reduces the amount of MHC class II transferred onto the T cells (82).
A very recent study related to the effect of varying cholesterol concentrations on GPI-linked and native I-Ek showed that the diffusion coefficients of both are dependent on cholesterol concentration. At a low cholesterol concentration, the diffusion coefficients are reduced up to a factor of 60 for native and 190 for GPI-linked I-Ek. The effect is reversed after cholesterol reintroduction (83). Although the lateral mobility of cell surface macromolecules under the parasitized condition has not been investigated to date, increased ruffling activity has been observed in Mφ at the site of phagosome formation during Leishmania parasite uptake (84). There is a report that Abs specific for a distinct CCR5 epitope lost their ability to bind to the cell surface after cholesterol extraction to varying degrees (85). The cause of the reduced ability of anti-CD48 and CTX-B to bind to P388D1(I) is not known, but it is tempting to speculate that there is a decrease in either their number or their accessibility or, perhaps, a combination of both. Previously, we have demonstrated that splenic Mφ of late stage-infected mice showed altered morphology, as evident from scanning electron microscopic studies (11). The cells were reported to undergo a change in shape after cholesterol depletion (83).
For both the initiation and the termination of the cognate immune response, the formation of an immunological synapse between T cells and APC is recognized as a key event (18, 19). The mature synapse lasts for several hours and is thought to be important for sustained signaling (86). Lipid rafts accumulate in the immunological synapse (21, 87). It has been shown in both Th cell blasts and CTLs that it is possible to detect even one peptide-MHC complex and that Ca2+ mobilization increased in a quantal manner up to the engagement of ∼10 agonist peptide-MHC complexes, which also occurs when a stable, mature, immunological synapse is formed (88). We have studied the ability of either peptide-pulsed or -unpulsed P388D1(N), P388D1(I), and L-P388D1(I) to form an immunological synapse with peptide-specific T cells. The extent of conjugate formation was 75, 15, and 77% with P388D1(N), P388D1(I), and L-P388D1(I), respectively. Because intracellular Ca2+ mobilization in the T cell in response to peptide-pulsed APCs is one of the important events of mature synapse formation, it is very likely that P388D1(I) is unable to form a peptide-dependent synapse. When P388D1(I) was used as an APC to drive T cells at a low peptide dose, there was less IL-2 production, which corroborated well with intracellular Ca2+ mobilization. This is in agreement with the reports on the association between Ca2+ mobilization and IL-2 production (89) and the initiation of synapse formation for T cells by the clustering of TCRs concomitant with a rise in intracellular Ca2+ levels (90, 91).
Collectively, our data show that LD infection of P388D1 cells results in a loss of Ag-presenting ability at a low peptide dose. The increase in fluidity after infection leads to a disruption of lipid rafts, which, in turn, affects Ag presentation at a low peptide dose. Thus, although the number of peptide-MHC complexes present on the infected cell surface is sufficient, they cannot possibly be aggregated in the raft to stimulate T cells. Correction of this increased fluidity, either physically by incubating cells below transition temperature or chemically by loading cholesterol, resulted in correction of defective APC function in LD-infected cells. As far as the ability of Ag presentation by cholesterol-loaded, infected APCs is concerned, the threshold of T cell activation can be further set by rafting MHC class II via concentrating ligands and thus promoting T cells for sensing low density Ag. Finally, our study has important implications for designing therapeutics against leishmaniasis. The delivery of antileishmanial drugs encapsulated in cholesterol-rich liposome may have additional beneficial effects.
We thank Drs. Damo Xu and Farrokh Modabber for critically reviewing the paper.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported by the Council for Scientific and Industrial Research, New Delhi, India (Project CMM002). D.C. and S.B. are recipients of Council of Scientific and Industrial research fellowships.
Abbreviations used in this paper: Mφ, macrophage; LD, Leishmania donovani; CTX-B, cholera toxin B subunit; DC, dendritic cell; DPH, 1,6-diphenyl-1,3,5-hexatriene; FA, fluorescence anisotropy; HEL, hen egg lysozyme; I-Mφ, infected splenic macrophage; KMP-11, kinetoplastid membrane protein-11; L-P388D1(I), cholesterol-loaded P388D1(I); P388D1(I), infected P388D1 cell; P388D1(N), uninfected P388D1 cell; SLA, soluble leishmanial Ag.